Tremendous progress has been made in computational protein design in the last 10 years. Though most work has focused on redesign of existing proteins to enhance their stability, specificity or new functionality, successful designs of novel protein folds (not naturally occurring) have emerged. However, to make protein design useful in practical applications, many challenges still exist in the design of protein-protein and protein- DNA interface and the design of dynamic properties of proteins, etc. To conquer those challenges, a practical solution is to develop more accurate scoring functions. It has been proved both experimentally and theoretically that intra and intermolecular signaling between distant sites within one or among many proteins plays a significant role in many biological processes, such as signal transmission and allosteric regulation, etc. The main research goal of this R21 grant is to develop new algorithms and strategies to incorporate the site-site couplings into a protein design procedure. Specifically, in Aim 1, we propose to study the site-site couplings and cooperative interactions of a protein or protein system using both the sequence-based and the physics-based approaches. Critical assessment of the existing approaches will be conducted and a novel algorithm will be developed using the residue-residue interaction energies. In Aim 2, the best methods for site-site couplings prediction will be tailored and integrated to physical scoring functions of protein design. To guarantee the success of this research grant, we propose four schemes to incorporate/combine the site-site coupling into a protein design procedure. The first two schemes represent simple combinations of sequence-based and physics-based protein design approaches; on the other hand, the next two schemes represent a more advanced integration of site-site correlations and physical scoring functions. In the last protocol, we propose to develop a novel scheme and software package to conduct protein design using the residue-residue interaction energies as a template. This approach, which is referred to as correlation embedding (CE), is based on a hypothesis that the correlation information is intrinsically encoded in the interaction energy matrix. The success of this sub-aim will have a great impact on rational protein design, as it represents a perfect integration of site-site couplings into a physical scoring function and opens a new avenue to conduct protein design. All the designed promising PDZ proteins will be synthesized and tested in both folding and functional assays.